Effect of zirconium on the microstructure and mechanical properties of long period stacking ordered Mg96Gd3Zn1 alloy

Effect of zirconium on the microstructure and mechanical properties of long period stacking ordered Mg96Gd3Zn1 alloy

Materials Science & Engineering A 560 (2013) 847–850 Contents lists available at SciVerse ScienceDirect Materials Science & Engineering A journal ho...

1MB Sizes 0 Downloads 6 Views

Materials Science & Engineering A 560 (2013) 847–850

Contents lists available at SciVerse ScienceDirect

Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea

Rapid communication

Effect of zirconium on the microstructure and mechanical properties of long period stacking ordered Mg96Gd3Zn1 alloy J.S. Zhang n, W.B. Zhang, X.Q. Ruan, L.P. Bian, W.L. Cheng, H.X. Wang, C.X. Xu College of Materials Science and Engineering, Taiyuan University of Technology, Taiyuan 030024, China

a r t i c l e i n f o

abstract

Article history: Received 20 June 2012 Received in revised form 6 October 2012 Accepted 8 October 2012 Available online 17 October 2012

The incorporation of Zr enhanced the mechanical properties of the long period stacking ordered (LPSO) structure Mg–Gd–Zn alloys. The reason was found to be the fact that Zr induced the precipitation of new phases, the variations in morphology and distribution of LPSO structure phases and the grain homogenization. & 2012 Elsevier B.V. All rights reserved.

Keywords: Mg–Gd–Zn alloys Long period stacking ordered (LPSO) Microstructure Mechanical properties

1. Introduction Strengthening magnesium (Mg) alloys is attracting ever increasing attention nowadays. The finding of LPSO structure in Mg–RE–Zn alloys, together with the addition of Zr, greatly enriched the ways to strengthen Mg alloys [1–8]. The Mg–Gd– Zn–Zr alloys have been obtained with a highest tensile strength (sb) about 290 MPa, a yield strength (s0.2) about 162 MPa and a elongation (d) about 10.35%, due to the appearance of LPSO phases [9]. However, to our best knowledge, the effect of Zr on the formation and transformation of LPSO structure phases has not been reported, which is of fundamental significance. In this study, the microstructure and mechanical properties of the alloys with and without Zr, Mg96  xGd3Zn1Zrx (at%) (x¼0 and 0.2) alloys, were systematically investigated, so as to make clear the strengthening mechanism of Zr in Mg–Gd–Zn alloys with LPSO structure.

2. Experimental procedures The raw material was melted and refined with an well type electric resistance furnace under a novel mixed protective gas atmosphere of CH2FCF3 (2.6 vol%) and N2 (97.4 vol%) at 1033 K. Then it was cast into a preheated mold. The solution-treating at 773 K for 50 h and the aging-treating at 473 K for 80 h were n

Corresponding author. Tel./fax: þ 86 351 601 8208. E-mail address: [email protected] (J.S. Zhang).

0921-5093/$ - see front matter & 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msea.2012.10.022

implemented in a SX2-8-10 box type high-temperature electric resistance furnace, followed by water quenching. The tensile tests of specimens with a gauge dimension of 60 mm  12 mm  2.5 mm were performed by a DNS100 electronic universal material testing machine with an initial strain rate of 3.33  10  4 s  1. Microhardness was measured by a HVS-1000 A Vikers hardness testing machine, with load of 10 g and loading time of 15 s. Brinell hardness was measured by a HB-3000 Brinell hardness testing machine with load of 62.5 kg and loading time of 15 s. Phase constitution analyses were performed with Y-2000 X-ray diffraction (XRD), using monochromatic Cu-Ka radiation. The microstructures and compositions of different phases in alloy were investigated by scanning electron microscopy (SEM, JSU-6700F) equipped with energy dispersive spectroscopy (EDS) and transmission electron microscopy (TEM, JEOL 2010). Thin foils for TEM observation were prepared by cutting the bulk sample into slices, grinding to the thickness of about 50 mm, and ion milling finally. 3. Results and discussion XRD patterns (Fig. 1) and EDS results (Fig. 2a) demonstrate that the as-cast microstructures of both alloy A (Fig. 3a) and B (Fig. 3b) consisted of a-Mg and (Mg, Zn)3Gd eutectic structure, which located at grain boundaries. In addition, the EDS results of the matrix of as-cast alloy A and B (Fig. 2b and c) show that a little Zr appeared in matrix after the addition of Zr. And the TEM analysis (Fig. 4a and b) proves the existence of 2H–Mg and 14HLPSO structures in matrix of both alloy A and B. The SEM images indicate that Zr induced the grain refinement and caused the

848

J.S. Zhang et al. / Materials Science & Engineering A 560 (2013) 847–850

a-Mg (Mg,Zn)3Gd LPSO phases Spherical phases

Element

Weight%

Mg K

21.27

Atomic% 55.90

Zn K

21.26

20.76

Gd L

57.47

23.34

Element

Weight %

Atomic %

Mg K

98.47

99.72

Zn K

0.19

0.07

Gd L

1.34

0.21

Element

Weight %

Atomic %

Mg K

97.70

99.52

Zn K

0.26

0.10

Gd L

1.52

0.24

Zr L

0.52

0.14

Element

Weight %

Atomic %

Mg K

9.24

25.82

Zn K

24.98

25.97

Intensity(a.u.)

as-cast

solid solution

aging-treated

20

30

40

50 2θ (degree)

60

70

80

Fig. 1. XRD patterns of as-cast, solid-solution and aging-treated alloy B.

transformation of the eutectic structure from dendrite shape (Fig. 3a) to homogeneous spherical shape (Fig. 3b). As a result, the stress generated during deformation of alloy can be absorbed by the homogeneous spherical grains, avoiding the stress concentration, thus contributing to the reinforcement. As shown in Fig. 3c, after solution-treating at 773 K for 50 h, eutectic structures at the grain boundary underwent a change in alloy A, where a kind of lamellar-type phase (with 14H-LPSO structure) reported by Yamasaki et al. [8,10] and Zeng et al. [9] transformed from eutectic structures. Nevertheless, massive phases could be found in solid-solution alloy B as shown in Fig. 3d, in the corresponding selected area electron diffraction (SAED) patterns (Fig. 4c), small periodic diffraction spots at the interval of 1/14 of distance between direct spot and (0002)Mg reflection were observed, the spot of (00014) corresponded to that of (0002)Mg. Based on the analyses mentioned above and the XRD peaks in Fig. 1, the massive phases were identified as 14HLPSO structure phases [8]. In addition, needle-like phases being determined as Mg25.82 Zn25.97Gd1.10Zr47.11 (at%) by EDS (Fig. 2d) were observed in the solid-solution alloy B and the subsequent aging-treated alloy B. Meanwhile, the Zr content in matrix of solid-solution alloy B (Fig. 2e) decreased compared to that of as-cast alloy B (Fig. 2c). Two kinds of lattices were observed in the SAED patterns in Fig. 4d, one was hexagonal close packed (hcp) lattice of Mg, another was face-centered cubic (fcc) of needle-like phases. To our best knowledge, needle-like phases were the first time to be reported. As seen in bright-field (BF) image (in Fig. 4d), high density dislocations were clearly observed around needle-like phases, indicating that these phases were effective to hinder the movement of dislocations when basal slip was occurred. Therefore, the improved mechanical properties could be partly ascribed to the occurrence of the needle-like phases, which will be discussed later. Fig. 3e shows that the LPSO structure phases in aging-treated alloy A nearly kept the same as that in solution-treating alloy A (Fig. 3c), they still showed lamellar shape and located at the grain boundaries. This was consistent with the result reported by Zeng et al. [9]. However, microstructure in the aging-treated alloy B changed a lot compared to not only the solution-treating alloy B but also the aging-treated alloy A. As seen in Fig. 3f, grain boundaries were clearly observed, but no lamellar LPSO phases were observed at the location of grain boundaries, instead, novel fine-stripy phases which owned the same orientation appeared

Zr L

63.24

47.11

Gd L

2.54

1.10

Element

Weight %

Atomic %

Mg K

99.00

99.79

Zn K

0.19

0.07

Gd L

0.71

0.11

Zr L

0.11

0.03

Fig. 2. EDS spectra for (a) eutectic structure, (b) matrix of as-cast alloy A, (c) matrix of as-cast alloy B, (d) needle-like phases and (e) matrix of solid-solution alloy B.

within grains. In addition, tiny spherical phases were observed at the interfaces of grains. The composition of the fine-stripy phases was determined as Mg86.62Zn4.25Gd9.13 (at%) by EDS. In the SAED patterns derived from the fine-stripy phases (Fig. 4e), the spots were arranged in positions that divided the height between the incident beam and the (0002) fundamental spot of the hcp cell 14-fold. Based on the XRD peaks in Fig. 1 and the analysis above, the fine-stripy phases were determined as 14H-LPSO structure phases [8]. Therefore, it can be concluded that LPSO phases in aging-treated alloy B kept consistent with solid-solution alloy B in LPSO types, but with change in morphology and distribution. Due to the variation in

J.S. Zhang et al. / Materials Science & Engineering A 560 (2013) 847–850

849

Dendrite Spherical grains

Massive LPSO phases Mg matrix Lamellar LPSO phases

Needle-like phases

Stripy LPSO phases Lamellar LPSO phases Spherical phases Mg matrix

Fig. 3. SEM images of (a) as-cast, (c) solid-solution and (e) aging-treated alloy A. (b) As-cast, (d) solid-solution and (f) aging-treated alloy B.

0002Mg 0002Mg

0000Mg

0000Mg 14H

14H

000,1414H 0001414H

2H

2H 000014H

000014H 000,1414H

C*

2110 2112

000014H

C*

Needle-like phases

111

0002

200

111

20nm

Dislocation

200

111

0002

C*

0002

LPSO phases Spherical phases

111 000

0000

14H

2H 14H

2H

14H

Dislocation 2H

Dislocation 20nm Fig. 4. Bright-field images and/or corresponding SAED patterns of (a) matrix of as-cast alloy A; (b) matrix of as-cast, (c) massive phases in solid-solution, (d) needle-like phases in solid-solution and aging-treated, (e) sandwich structures in aging-treated and (f) spherical phases in aging-treated alloy B.

850

J.S. Zhang et al. / Materials Science & Engineering A 560 (2013) 847–850

Table 1 Mechanical properties of alloy A and B under different heat treatment states.

350

Stress / MPa

300

Alloy

Composition (at%)

Grain size/mm

As-cast alloy A Aging alloy A As-cast alloy B Aging alloy B

Mg96Gd3Zn1

28

Mg96Gd3Zn1

aging-treated alloy B

250

aging-treated alloy A 200 as-cast alloy B

150

as-cast alloy A

100

Brinell Hardness/HB

UTS/ MPa

Elongation/ %

81

174

4.5

31

103

279

5.3

Mg95.8Gd3Zn1Zr0.2 20

83

223

4.9

Mg95.8Gd3Zn1Zr0.2 21

124

332

6.9

50 0 0

1

2

3

4 5 Strain / %

6

7

8

Fig. 5. Tensile stress–strain curves of as-cast alloy A and B, aging-treated alloy A and B.

distribution, the LPSO phases with high strength and toughness uniformly dispersed in alloys, which contributed to increase of strength and ductility. In addition, because of the novel morphology of LPSO phases, sandwich structures [11] consisting of 2H–Mg and 14H-LPSO (in Fig. 4c) distributed much more widely in agingtreated alloy B, the fine-stripy LPSO phases acted as obstacles that resisted the movement of dislocations containing in the 2H–Mg. In order to identify the spherical phases, their composition was identified as Mg58.90Zn13.76Gd27.34 (at%) by EDS, and the lattice structure was identified as fcc by SAED patterns (in Fig. 4f). Besides, the new peaks compared to the solid-solution alloy B in XRD patterns also prove the existence of the new phases which were named as Mg4ZnGd2 phases. The Mg4ZnGd2 phases with a average size of about 2 mm owned a microhardness of 160 HV, much higher than the average microhardness of matrix (110 HV). Moreover, they precipitated at the interfaces of several adjacent grains. So it is easy to assume that the Mg4ZnGd2 phases played an important role in resisting the movement of grain boundaries during deformation. Besides, Mg4ZnGd2 phases were also observed to hinder the slipping of fine-stripy LPSO phases and dislocations as seen in Fig. 4f. The analyses mentioned above indicated that the spherical Mg4ZnGd2 phases contributed to the enhancement of the mechanical properties. Typical stress–strain curves from tensile tests are shown in Fig. 5. Related mechanical properties are summarized in Table 1. It can be seen that the aging-treated alloy A and B had better mechanical properties than as-cast alloy A and B respectively due to the formation of LPSO phases after heat treatment. Furthermore, agingtreated Zr-containing alloy B showed best mechanical properties, especially the tensile strength, yield strength and Brinell hardness were superior to that of un-extruded Mg–Gd–Zn–Zr alloys which have been reported [8,9,12]. The reason can be summarized as that Zr induced the grain homogenization and refinement, avoiding the stress concentration during deformation. Zr promoted the variation in the morphology and distribution of LPSO phases, achieving the uniform distribution of LPSO phases with high strength and toughness, and forming the sandwich structure which limited the

movement of dislocations in 2H–Mg. When basal slip was activated, needle-like phases, with the orientation parallel to the c-axis, hindered the movement of dislocations. Spherical phases with high hardness resisted the movement of grain boundaries, stripy LPSO phases and dislocations.

4. Conclusions Mg96 xGd3Zn1Zrx (at%) (x¼0 and 0.2) alloys were developed by the conventional permanent mold casting method. With 0.2 at% Zr addition, the as-cast alloy showed fine (20 mm) and homogeneous grains. After solution treatment, block LPSO phases were transformed from eutectic structures while conventional lamellar LPSO phases formed in Zr-free alloy. In Mg95.8Gd3Zn1Zr0.2 alloy, needlelike phases generated during solution treatment and still existed after aging treatment. Spherical phases with microhardness of 160 HV and novel fine-stripy LPSO phases which distributed dispersively in grains generated during aging treatment. The agingtreated Mg95.8Gd3Zn1Zr0.2 alloy exhibited optimal mechanical properties (UTS¼332 MPa, YS¼205 MPa, Elongation¼6.9% and Brinell Hardness¼124 HB) due to the formation of needle-like phases, spherical phases and the fine-stripy LPSO phases after the 0.2 at% Zr addition.

Acknowledgements The authors wish to acknowledge the financial support of the National Science Foundation of China (No. 50571073), Ph.D. Programs Foundation of Ministry of Education of China (20111402110004) and Natural Science Foundation of Shanxi Province (No. 2009011028-3). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]

D.D. Yin, Q.D. Wang, et al., Mater. Sci. Eng. A 546 (2012) 239–247. G.L. Bi, D.Q. Fang, et al., Mater. Sci. Eng. A 528 (2011) 3609–3614. J.S. Zhang, J. Yan, et al., Mater. Lett. 62 (2008) 4489–4491. K. Hagihara, A. Kinoshita, et al., Acta Mater. 58 (2010) 6282–6293. T. Honma, T. Ohkubo, et al., Acta Mater. 55 (2007) 4137–4150. G. Garce´s, E. On´orbe, et al., Mater. Sci. Eng. A 539 (2012) 48–55. Y.B. Chun, J. Geng, et al., Mater. Sci. Eng. A 528 (2011) 3653–3658. M. Yamasaki, M. Sasaki, et al., Acta Mater. 55 (2007) 6798–6805. X.Q. Zeng, Y.J. Wu, et al., Acta Metall. Sin. 46 (2010) 1041–1046. M. Yamasaki, Y. Kawamura, Scr. Mater. 60 (2009) 264–267. W.J. Ding, Y.J. Wu, et al., J. Mater. Res. 24 (2009) 1842–1853. K. Liu, J.H. Zhang, et al., Mater. Sci. Eng. A 505 (2009) 13–19.